Monoclonal antibodies to human tumor-associated differentiation antigens offer promises for the "targeting" of various antitumor agents such as radioisotopes, chemotherapeutic drugs, and toxins. [Order Baldwin and Byers, (eds.), in "Monoclonal Antibodies for Cancer Detection and Therapy", London, Academic Press (1985)]. In addition, some monoclonal antibodies have the advantage of killing tumor cells via ADCC or CDC in the presence of human effector cells or serum [Hellstrom et al., Proc. Natl. Acad. Sci. USA 83:7059-7063 (1986)], and there are a few monoclonal antibodies that have a direct antitumor activity which does not depend on any host component [Drebin et al., Oncogene 2:387-394 (1988)].
Many monoclonal antibodies reactive with carcinomaassociated antigens are known [see, e.g., Papsidero, "Recent Progress In The Immunological Monitoring Of Carcinomas Using Monoclonal Antibodies, Semin. Surg. Oncol., 1 (4):171-81 (1985); Schlom et al., "Potential Clinical Utility Of Monoclonal Antibodies In The Management Of Human Carcinomas", Important Adv. Oncol., 170-92 (1985); Allum et al., "Monoclonal Antibodies In The Diagnosis And Treatment of Malignant Conditions", Surg. Ann., 18:41-64 (1986); and Houghton et al., "Monoclonal Antibodies: Potential Applications To The Treatment Of Cancer", Semin. Oncol., 13(2):165-79 (1986)].
These known monoclonal antibodies can bind to a variety of different carcinoma-associated antigens including glycoproteins, glycolipids and mucins [see, e.g., Fink et al., "Monoclonal Antibodies As Diagnostic Reagents for The Identification And Characterization Of Human Tumor Antigens", Prog. Clin. Pathol., 9:121-33 (1984)]. For example, monoclonal antibodies that bind to glycoprotein antigens on specific types of carcinomas include those described in U.S. Pat. 4,737,579 (monoclonal antibodies to non-small cell lung carcinomas), U.S. Pat. No. 4,753,894 (monoclonal antibodies to human breast cancer), U.S. Pat. No. 4,579,827 (monoclonal antibodies to human gastrointestinal cancer), and U.S. Pat. No. 4,713,352 (monoclonal antibodies to human renal carcinoma). Monoclonal antibody B72.3, which is one of the antibodies studied the most, recognizes a tumor-associated mucin antigen of greater than 1,000 kd molecular weight that is selectively expressed on a number of different carcinomas. Thus, B72.3 has been shown to react with 84% of breast carcinomas, 94% of colon carcinomas, 100% of ovarian carcinomas and 96% of non-small cell lung carcinomas [see Johnston, "Applications of Monoclonal Antibodies In Clinical Cytology As Exemplified By Studies With Monoclonal Antibody B72.3", Acta Cytol., 1(5): 537-56 (1987) and U.S. Pat. No. 4,612,282, issued to Schlom et al.]. Another patented monoclonal antibody, KC-4, [see U.S. Pat. No. 4,708,930], recognizes an approximately 400-500 kd protein antigen expressed on a number of carcinomas, such as colon, prostate, lung and breast carcinoma. It appears that neither the B72.3 nor KC-4 antibodies internalize within the carcinoma cells with which they react.
Monoclonal antibodies reactive with glycolipid antigens associated with tumor cells have been disclosed. For example, Young et al., "Production Of Monoclonal Antibodies Specific For Two Distinct Steric Portions Of The Glycolipid Ganglio-N-Triosylceramide (Asialo GM.sub.2)", J. Exp. Med., 150: 1008-1019 (1979) disclose the production of two monoclonal antibodies specific for asialo GM.sub.2, a cell surface glycosphingolipid antigen that was established as a marker for BALB/c V3T3 cells transformed by Kirsten murine sarcoma virus. See, also, Kniep et al., "Gangliotriaosylceramide (Asialo GM.sub.2) A Glycosphingolipid Marker For Cell Lines Derived From Patients With Hodgkin's Disease", J. Immunol., 131(3): 1591-94 (1983) and U.S. Pat. No. 4,507,391 (monoclonal antibody to human melanoma).
Other monoclonal antibodies reactive with glycolipid antigens on carcinoma cells include those described by Rosen et al., "Analysis Of Human Small Cell Lung Cancer Differentiation Antigens Using A Panel Of Rat Monoclonal Antibodies", Cancer Research, 44:2052-61 (1984) (monoclonal antibodies to human small cell lung cancer), Varki et al., "Antigens Associated with a Human Lung Adenocarcinoma Defined by Monoclonal Antibodies", Cancer Research 44:681-87 (1984); (monoclonal antibodies to human adenocarcinomas of the lung, stomach and colon and melanoma), and U.S. Pat. No. 4,579,827 (monoclonal antibodies to human colon adenocarcinoma). See, also, Hellstrom et al., "Antitumor Effects Of L6, An IgG2a Antibody That Reacts With Most Human Carcinomas", Proc Natl Acad Sci, USA, 83:7059-63 (1986) which describes the L6 monoclonal antibody that recognizes a carbohydrate antigen expressed on the surface of human non-small cell lung carcinomas, breast carcinomas and colon carcinomas.
Additional monoclonal antibodies exhibiting a high specific reactivity to the majority of cells from a wide range of carcinomas are greatly needed. This is so because of the antigenic heterogeneity of many carcinomas which often necessitates, in diagnosis or therapy, the use of a number of different monoclonal antibodies to the same tumor mass. There is a further need, especially for therapy, for so called "internalizing" antibodies, i.e., antibodies that are easily taken up by the tumor cells to which they bind. Antibodies of this type find use in therapeutic methods utilizing antibody-drug or antibody-toxin conjugates wherein a therapeutic antitumor agent is linked to an antibody for delivery to the tumor, where the antibody binds to the tumor-associated antigen with which it is reactive and "delivers" the antitumor agent inside the tumor cells [see, e.g., Embleton et al., "Antibody Targeting Of Anti-Cancer Agents", in Monoclonal Antibodies For Cancer Detection and Therapy, pp. 317-44 (Academic Press, 1985)]. Antibodies to tumor-associated antigens which are not able to internalize within the tumor cells to which they bind are generally not useful to prepare conjugates with antitumor drugs or toxins, since these would not be able to reach their site of action within the cell. Other approaches would then be needed so as to use such antibodies therapeutically.
Several internalizing antibodies reacting with lymphocyte antigens are known. In contrast, such antibodies are rare when dealing with solid tumors. One of the few examples of an internalizing antibody reacting with carcinomas is an antibody disclosed in Domingo et al., "Transferrin Receptor As A Target For Antibody-Drug Conjugates" Methods Enzymol. 112:238-47 (1985). This antibody is reactive with the human transferrin-receptor glycoprotein expressed on tumor cells. However, because the transferrin-receptor is also expressed on many normal tissues, and often at high levels, the use of an anti-transferrin-receptor antibody in a antibody-drug or antibody-toxin conjugate may have significant toxic effects on normal cells. The utility of this antibody for selective killing or inhibition of tumor cells is therefore questionable. Another internalizing antibody is BR64 (disclosed in patent applications U.S. Ser. No. 289,635, filed Dec. 22, 1988, (now abandoned), and Ser. No. 443,696 filed Nov. 29, 1989, (now U.S. Pat. No. 5,242,824), and incorporated by reference herein), which binds to a large spectrum of human carcinomas.
The cell fusion technique for the production of monoclonal antibodies [Kohler and Milstein, Nature (London) 256:495 (1975)] has permitted the development of a number of murine monoclonal antibodies reactive with antigens, including previously unknown antigens. Murine monoclonal antibodies, however, may be recognized as foreign substances by the human immune system and neutralized such that their potential in human therapy is not realized. Therefore, recent efforts have focused on the production of so-called "chimeric" antibodies by the introduction of DNA into mammalian cells to obtain expression of immunoglobulin genes [Oi et al., Proc. Natl. Acad. Sci. USA 80:825 (1983); Potter et al., Proc. Natl. Acad. Sci. USA 81:7161; Morrison et al., Proc. Natl. Acad. Sci. USA 81:6581 (1984); Sahagan et al., J. Immunol. 137:1066 (1986); Sun et al., Proc. Natl. Acad. Sci. 84:214 (1987)].
Chimeric antibodies are immunoglobulin molecules comprising a human and non-human portion. More specifically, the antigen combining region (variable region) of a chimeric antibody is derived from a non-human source (e.g. murine) and the constant region of the chimeric antibody which confers biological effector function to the immunoglobulin is derived from a human source. The chimeric antibody should have the antigen binding specificity of the non-human antibody molecule and the effector function conferred by the human antibody molecule.
In general, the procedures used to produce chimeric antibodies involve the following steps:
a) identifying and cloning the correct gene segment encoding the antigen binding portion of the antibody molecule; this gene segment (known as the VDJ, variable, diversity and joining regions for heavy chains or VJ, variable, joining regions for light chains or simply as the V or variable region) may be in either the cDNA or genomic form; PA1 b) cloning the gene segments encoding the constant region or desired part thereof; PA1 c) ligating the variable region with the constant region so that the complete chimeric antibody is encoded in a transcribable and translatable form; PA1 d) ligating this construct into a vector containing a selectable marker and gene control regions such as promoters, enhancers and poly(A) addition signals; PA1 e) amplifying this construct in bacteria; PA1 f) introducing this DNA into eukaryotic cells (transfection) most often mammalian lymphocytes; PA1 g) selecting for cells expressing the selectable marker; PA1 h) screening for cells expressing the desired chimeric antibody; and PA1 k) testing the antibody for appropriate binding specificity and effector functions.
Antibodies of several distinct antigen binding specificities have been manipulated by these protocols to produce chimeric proteins [e.g. anti-TNP: Boulianne et al., Nature 312:643 (1984); and anti-tumor antigens: Sahagan et al., J. Immunol. 137:1066 (1986)]. Likewise, several different effector functions have been achieved by linking new sequences to those encoding the antigen binding region. Some of these include enzymes [Neuberger et al., Nature 312:604 (1984)], immunoglobulin constant regions from another species and constant regions of another immunoglobulin chain (Sharon et al., Nature 309:364 (1984); Tan et al., J. Immunol. 135:3565-3567 (1985)].
The discovery of homologous recombination in mammalian cells permits the targeting of new sequences to specific chromosomal loci. Homologous recombination occurs when cultured mammalian cells integrate exogenous DNA into chromosomal DNA at the chromosome location which contains sequences homologous to the plasmid sequences [Folger et al., Mol. Cell. Biol. 2:1372-1387 (1982); Folger et al., Symp. Quant. Biol. 49:123-138 (1984); Kucherlapati et al., Proc. Natl. Acad. Sci. USA 81:3153-3157 (1984); Lin et al., Proc. Natl. Acad. Sci. USA 82:1391-1395 (1985); de Saint Vincent et al., Proc. Natl. Acad. Sci. USA 80:2002-2006 (1983); Shaul et al., Proc. Natl. Acad. Sci. USA 82:3781-3784 (1985)]. The potential for homologous recombination within cells permits the modification of endogenous genes in situ. Conditions have been found where the chromosomal sequence can be modified by introducing into the cell a plasmid DNA which contains a segment of DNA homologous to the target locus and a segment of new sequences with the desired modification [Thomas et al., Cell 44:419-428 (1986); Smithies et al., Nature 317:230-234 (1985); Smith et al., Symp. Quant. Biol. 49:171-181 (1984)]. Homologus recombination between mammalian cell chromosomal DNA and the exogenous plasmid DNA can result in the integration of the plasmid or in the replacement of some of the chromosomal sequences with homologous plasmid sequences. This can result in placing a desired new sequence at the endogenous target locus.
The process of homologous recombination has been evaluated using genes which offer dominant selection such as NEO and HPRT for a few cell types [Song et al., Proc. Natl. Acad. Sci. USA 84:6820-6824 (1987); Rubinitz and Subramani, Mol. Cell Biol. 6:1608-1614 (1986); and Liskay, Cell 35:157-164 (1983)]. Recently, procedures for modifying antibody molecules and for producing chimeric antibody molecules using homologous recombination to target gene modification have been described [Fell et al., Proc. Natl. Acad. Sci. USA 86:8507-8511 (1989); and co-pending U.S. patent applications Ser. No. 243,873 filed Sep. 14, 1988, and Ser. No. 468,035 filed Jan. 22, 1990, assigned to the same assignee as the present application, all of which are incorporated by reference herein].
The most direct way to apply antitumor monoclonal antibodies clinically is to administer them in unmodified form, using monoclonal antibodies which display antitumor activity in vitro and in animal models. Most monoclonal antibodies to tumor antigens do not appear to have any antitumor activity by themselves, but certain monoclonal antibodies are known which mediate complement-dependent cytotoxicity (CDC), i.e. kill human tumor cells in the presence of human serum as a source of complement [see, e.g. Hellstrom et al., Proc. Natl. Acad. Sci. USA 82:1499-1502 (1985)], or antibody-dependent cellular cytotoxicity (ADCC) together with effector cells such as human NK cells or macrophages. To detect ADCC and CDC activity monoclonal antibodies are tested for lysing cultured .sup.51 Cr-labeled tumor target cells over a 4-hour incubation period.
Target cells are labeled with .sup.51 Cr and then exposed for 4 hours to a combination of effector cells (in the form of human lymphocytes purified by the use of a lymphocyte-separation medium) and antibody, which is added in concentrations varying between 0.1 .mu.g/ml and 10 .mu.g/ml. The release of .sup.51 Cr from the target cells is measured as evidence of tumor-cell lysis (cytotoxicity). Controls include the incubation of target cells alone or with either lymphocytes or monoclonal antibody separately. The total amount of .sup.51 Cr that can be released is measured and ADCC is calculated as the percent killing of target cells observed with monoclonal antibody plus effector cells as compared to target cells being incubated alone. The procedure for CDC is identical to the one used to detect ADCC except that human serum, as a source of complement, (diluted 1:3 to 1:6) is added in place of the effector cells.
Monoclonal antibodies with ADCC and CDC activity are considered for therapeutic use because they often have anti-tumor activities in vivo. Antibodies lacking ADCC and CDC activity in vitro, on the other hand, are commonly ineffective in vivo unless used as carriers of antitumor agents. The ability of a monoclonal antibody to activate the host's complement may prove to be therapeutically beneficial not only because tumor cells may be killed, but also because the blood supply to tumors may increase, thus facilitating the uptake of drugs [see Hellstrom et al., "Immunological Approaches to Tumor Therapy: Monoclonal Antibodies, Tumor Vaccines, and Anti-Idiotypes, in Covalently Modified Antigens and Antibodies in Diagnosis and Therapy, Quash & Rodwell, eds., Marcel Dekker, pp. 15-18 (1989)]. Among mouse monoclonal antibodies, the IgG2a and IgG3 isotypes are most commonly associated with ADCC and CDC. Antibodies having both ADCC and CDC activity have high selectivity for killing only the tumor cells to which they bind and would be unlikely to lead to toxic effects if nonspecifically trapped in lung, liver or other organs. This may give such antibodies an advantage over radiolabelled antibodies or certain types of immunoconjugates.
Very few antibodies are able to kill tumor cells by themselves, that is, in the absence of effector cells or complement as in ADCC or CDC. BR96 is such an antibody, because it can kill cells by itself at an antibody concentration of approximately 10 .mu.g/ml or higher. Such antibodies are of particular interest since they can interfere with some key event in the survival of neoplastic cells.
It is thus apparent that an antibody that displays a high degree of selectivity to a wide range of carcinomas, has anti-tumor activity by itself, and is capable of being readily internalized by tumor cells, may be of great benefit in tumor therapy.